GitLab

Revert of Restrict ModalDialogTest#testPromptModalDialog to non-low-end devices (patchset #1 id:1 of https://codereview.chromium.org/2397573008/ )

Reason for revert:
Root cause is in Android framework, we're just going to catch the exception

Original issue's description:
> Restrict ModalDialogTest#testPromptModalDialog to non-low-end devices
>
> BUG=653685
> TBR=carlosk@chromium.org, dfalcantara@chromium.org
>
> Committed: https://crrev.com/3988c1fd1bc405d01850a8e5e28619d0630b5184
> Cr-Commit-Position: refs/heads/master@{#423709}

TBR=carlosk@chromium.org,dfalcantara@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=653685

Review-Url: https://codereview.chromium.org/2396243002
Cr-Commit-Position: refs/heads/master@{#423743}

When an IndexedDB versionchange transaction is aborted, Blink's
IndexedDB metadata is not entirely reverted, causing our behavior to
deviate from the IndexedDB specification in minor, subtle ways. This
change aligns our behavior with the spec and with Firefox's
implementation.

The change also (slightly) reduces IndexedDB memory usage in two ways.
(1) Object store and index metadata is now shared between an IDBDatabase
and its associated IDBObjectStore and IDBIndex instances, instead of
being copied. (2) versionchange transactions only back up the metadata
for object stores that are accessed by JavaScript, instead of creating a
backup for the entire database metadata.

BUG=645018,457447

Review-Url: https://codereview.chromium.org/2314933005
Cr-Commit-Position: refs/heads/master@{#423742}

This is not longer needed after we switch to using offline interceptor

BUG=
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2395763003
Cr-Commit-Position: refs/heads/master@{#423741}

This was deemed unnecessary in UX review. When navigator.share is used
from incognito, you still see the intent picker which is considered
sufficient warning for users.

The bulk of the code to support this (https://crrev.com/420564) remains
and will be removed in a future CL.

BUG=645007

Review-Url: https://codereview.chromium.org/2389393004
Cr-Commit-Position: refs/heads/master@{#423740}

This adds an MB config for 'Linux ChromeOS Buildspec Tests' builder, and makes the 'linux_packages_all' target be defined for CrOS builds as well as desktop Linux, though I have no idea if this'll actually build in that config.

TBR=dnj@chromium.org, brettw@chromium.org
BUG=643238

Review-Url: https://codereview.chromium.org/2399883003
Cr-Commit-Position: refs/heads/master@{#423739}

After a sequence of frames that don't saturate bandwidth libvpx always
chooses lowest allowed quntizer (highest quality). As result these frames
are quite big after being encoded, which results in poor response latency
for those frames. With this change the scheduler detects those frames and
sets min_quantizer to 60 to ensure that they are first encoded with low
quality and quality is topped off later.

This change reduces latency for "big" frames in
ProtocolPerfTest.TotalLatencyWebrtc from 500ms to 200ms in 8Mbps case,
even without proper BW estimation.

Also changed frame duration from 66 to 33 ms, to match actual frame
duration (this doesn't seem to have significant effect on quality).

BUG=645656

Review-Url: https://codereview.chromium.org/2381213002
Cr-Commit-Position: refs/heads/master@{#423738}

Chrome now uses the browser process for background download resumption.
We don't need a separate process for this, remove the code.

TBR=palmer@chromium.org

Review-Url: https://codereview.chromium.org/2365093002
Cr-Commit-Position: refs/heads/master@{#423737}

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/b5cecfe7c5a9684ea21d08599cbcc85a419dd190 Roll recipe dependencies (trivial). (recipe-roller@chromium.org)
depot_tools:
  https://crrev.com/a3f10315eb5597be73cdfc1a4a1145fe3cc3b541 Roll recipe dependencies (trivial). (recipe-roller@chromium.org)
recipe_engine:
  https://crrev.com/825386b5d616ed1e5fca94ec76eb6de9335b86d2 Add astunparse and update_vendoring. (iannucci@chromium.org)

TBR=martiniss@chromium.org,phajdan.jr@chromium.org
BUG=

Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Review-Url: https://codereview.chromium.org/2400843002
Cr-Commit-Position: refs/heads/master@{#423736}

This CHECK is hit by MSE users who make a small Remove() of their last
appended frames and then a subsequent Append() as if the previous remove
had not occurred.

The bug will also manifest if users set duration to just under the
current duration - currently equivalent to a small Remove().
Aside: duration truncation will soon be deprecated.

If the initial Remove() is small enough (less than the fudge room),
RemoveInternal() will fail to see this as causing a discontinuity and will
not reset range_for_next_append_. If the next append is beyond the
newly-shortened-range-end + fudge room, then CanAppendToEnd will CHECK.

BUG=627905
TEST=new unit tests, crash gone from (nsfw) http://ck101.com/thread-3559838-1-1.html?ref=banner

Review-Url: https://codereview.chromium.org/2385423002
Cr-Commit-Position: refs/heads/master@{#423735}

SERP will dispatch intents directly to instant apps. If the user clicked
on http/https link on SERP, it should not offer or launch instant apps.

BUG=644772

Review-Url: https://codereview.chromium.org/2397153003
Cr-Commit-Position: refs/heads/master@{#423734}

SyncExtensionFunctions, AsyncExtensionFunctions,
ChromeSyncExtensionFunctions, and ChromeAsyncExtensionFunctions are
deprecated.

Remove ChromeSyncExtensionFunctions from:
- tabs

BUG=634140

Review-Url: https://codereview.chromium.org/2398793002
Cr-Commit-Position: refs/heads/master@{#423733}

A couple triggers crashes, which I've filed separately. The keyboard
ones won't work in mash. I plan to move those out of this file
separately.

BUG=624172,652871
TEST=covered by tests
R=jamescook@chromium.org

Review-Url: https://codereview.chromium.org/2391153002
Cr-Commit-Position: refs/heads/master@{#423732}

BUG=593596

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2238883006
Cr-Commit-Position: refs/heads/master@{#423731}

Every implemented function of the Adapter Mojo service requires a
reference to the system's Bluetooth adapter. Since any function could be called
at any time in the service, the adapter must be available from the creation of
the Adapter service. To satisfy this requirement, a factory pattern has been
implemented to create instances of the Adapter service with the required
reference to a Bluetooth adapter.
BUG=651282
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2379573006
Cr-Commit-Position: refs/heads/master@{#423730}

It is bad user experience to show sign in promotions to users who
eventually cannot sign in anyway, because of missing gmscore. This CL
suppresses sign in promotions by checking the availability of gmscore.

BUG=653615

Review-Url: https://codereview.chromium.org/2399183002
Cr-Commit-Position: refs/heads/master@{#423729}

Currently, w3c-test-autoroller is blocked on importing
because new baseline files are not added and committed
before uploading.  Example of a build where this happens:
https://build.chromium.org/p/chromium.infra.cron/builders/w3c-test-autoroller/builds/7732.

Purpose: This CL would unblock this issue on w3c-test-autoroller.

This would be unnecessary (and could be reverted) if either of
http://crrev.com/2397573002 or http://crrev.com/2396433004 is committed.

BUG=621599

Review-Url: https://codereview.chromium.org/2394913002
Cr-Commit-Position: refs/heads/master@{#423728}

Move tests to depends on the LayerTreeHost, instead of the InProcess
subclass, in preparation of setting up the framework to have tests
run in remote mode.

There is still an accessor for LayerTreeHostInProcess, only it DCHECKS
that the test is running in threaded or single-threaded mode.

BUG=653371
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2400733002
Cr-Commit-Position: refs/heads/master@{#423727}

Currently we only send device properties and advertising data to
Android only when we first found a new device. We should also
update Android when we got DeviceChanged event.

Also refactor the code to
- Add new SendDevice() for common code related to this.
- Filter invalid data out from data that will send to Android.

BUG=653310,b:31916649
TEST=nRF Connect app in minnie correctly show rssi graph

Review-Url: https://codereview.chromium.org/2391973003
Cr-Commit-Position: refs/heads/master@{#423726}

Disable View System hardware acceleration to see if users on this specific device
no longer crash in M54.

BUG=651918

Review-Url: https://codereview.chromium.org/2397903003
Cr-Commit-Position: refs/heads/master@{#423725}

BUG=650852

Review-Url: https://codereview.chromium.org/2387333006
Cr-Commit-Position: refs/heads/master@{#423724}

This CL should not introduce any change in behavior. The code should now
look and behave as if the Overview Mode has originally been implemented
with the Material Design look and feel.

BUG=604767, 614453
TEST=WindowSelectorTest

Review-Url: https://codereview.chromium.org/2401473002
Cr-Commit-Position: refs/heads/master@{#423723}

Move some of the code into two new functions to make everything more readable.

R=eae@chromium.org,glebl@chromium.org
BUG=635619

Review-Url: https://codereview.chromium.org/2400023002
Cr-Commit-Position: refs/heads/master@{#423722}

This allows multiple client proxies in a render process to connect
to a single BrowserAssociatedInterface at the same time.

BUG=612500

Review-Url: https://codereview.chromium.org/2400663003
Cr-Commit-Position: refs/heads/master@{#423721}

If the gpu process crashes the worker context will return null sync
tokens which the resource write lock ignores. It's possible for the
compositor context to not detect the gpu process crash until much later.
Until then the scheduler might initiate a draw which will crash when
DCHECKing that resources being sent have sync tokens.

This CL also enables sync tokens for LayerTreeTests by default and
includes some minor cleanup in LayerTreeHostTestContext.

R=brianderson@chromium.org,danakj@chromium.org
BUG=645736
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2387333003
Cr-Commit-Position: refs/heads/master@{#423720}

BUG=653701
TBR=mlamouri@chromium.org

Review-Url: https://codereview.chromium.org/2396223002
Cr-Commit-Position: refs/heads/master@{#423719}

[Bootstrapping] Introduce an accelerator (Ctrl+Alt+Shift+S) to put a ChromeOS device into Slave OOBE process.

In order to enhance the security and avoid the Bluetooth in-middle attach between
Master and Slave, we've decided to do the followings for a Slave device:
- For devices lacking sufficient input, e.g., Chromebit, keep the state as it is today:
  - The ChromeOS device will put itself in Bluetooth discoverable mode in the first
    OOBE screen (HID detection screen) and accpet the incoming connection request
    from the Bootstrapping app.
- For devices with sufficient input, e.g., Chromebook:
  - The ChromeOS device will have Bluetooth disabled by default as in normal OOBE flow
  - Upon the user pressing the accelerator (Ctrl+Alt+Shift+S) in the first OOBE screen
    (Network screen): the device will put itself in Bluetooth discoveralbe mode and
    accept the incoming connection request from the Bootstrapping app.

BUG=652003

Review-Url: https://codereview.chromium.org/2383953002
Cr-Commit-Position: refs/heads/master@{#423718}

This method is never used and is unimplemented on most platforms.

BUG=599570

Review-Url: https://codereview.chromium.org/2399993002
Cr-Commit-Position: refs/heads/master@{#423717}

Use short version in RelaunchAppWithHelper() instead of
[[NSRunningApplication currentApplication] isActive]

BUG=650854

Review-Url: https://codereview.chromium.org/2398943002
Cr-Commit-Position: refs/heads/master@{#423716}

1. Both OK and "Load full site" buttons should be present.
2. Infobar should have message and explanation text

BUG=609747

Review-Url: https://codereview.chromium.org/2393813002
Cr-Commit-Position: refs/heads/master@{#423715}

BUG=653609
TBR=qinmin@chromium.org

Review-Url: https://codereview.chromium.org/2402533002
Cr-Commit-Position: refs/heads/master@{#423714}

We should not do anything for the spanner placeholder but just walk the
spanner directly.

Out-of-flow positioned descendants of a multicol spanner need to be
specially handled because their container may be not their ancestor in
the order of PrePaintTreeWalk.

This fixes the following layout tests for slimmingPaintInvalidation:
fast/multicol/dynamic/*spanner*.html
fast/multicol/span/*.html
(except for fast/multicol/dynamic/abspos-multicol-with-spanner-becomes-spanner.html
which is still failing because of some other reason.)

BUG=646176
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2388723004
Cr-Commit-Position: refs/heads/master@{#423713}

R=thakis@chromium.org
BUG=563793

Review-Url: https://codereview.chromium.org/2396433006
Cr-Commit-Position: refs/heads/master@{#423712}

BUG=647441
TEST=manual

Review-Url: https://codereview.chromium.org/2338173005
Cr-Commit-Position: refs/heads/master@{#423711}

12 kb may be too small for a 64-bit machine. This allows enough
breathing space for the thread local storage teardown later on.

BUG=590907

Review-Url: https://codereview.chromium.org/2395303002
Cr-Commit-Position: refs/heads/master@{#423710}

BUG=653685
TBR=carlosk@chromium.org, dfalcantara@chromium.org

Review-Url: https://codereview.chromium.org/2397573008
Cr-Commit-Position: refs/heads/master@{#423709}

Also drop an unused include of ReferenceFilterBuilder.h.

BUG=439970
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2400663002
Cr-Commit-Position: refs/heads/master@{#423708}

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/fecd39cfc57157898353120c65b5d3e054451cb8 Remove "Precise" from chromium.fyi builder names (jeffcarp@chromium.org)

TBR=martiniss@chromium.org,phajdan.jr@chromium.org
BUG=652691

Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Review-Url: https://codereview.chromium.org/2401773002
Cr-Commit-Position: refs/heads/master@{#423707}

This is only used in URLRequestFtpJob.

R=mmenke@chromium.org
BUG=529319

Review-Url: https://codereview.chromium.org/2398153002
Cr-Commit-Position: refs/heads/master@{#423706}

This is a work-around for flickering in CCT for M54. Increasing the
max number of saved frames for high-end devices (>= 3.5 GB memory)
will prevent a flash that comes from bringing a background tab into
the foreground whose frame has been evicted.

BUG=653249

Review-Url: https://codereview.chromium.org/2394113003
Cr-Commit-Position: refs/heads/master@{#423705}

This test fails consistently on low-end Android devices.
Return early instead of running the test.

BUG=653637
TBR=eseckler@chromium.org, dgozman@chromium.org

Review-Url: https://codereview.chromium.org/2395863004
Cr-Commit-Position: refs/heads/master@{#423704}